Re: A common problem with SLAAC in "renumbering" scenarios

[Alexandre Petrescu <[e-mail]>]

Le 13/02/2019 à 15:59, Richard Patterson a écrit :
> On Wed, 13 Feb 2019 at 14:50, Alexandre Petrescu
> <[e-mail]> wrote:
> 
>> Le 05/02/2019 à 11:31, Mikael Abrahamsson a écrit :
>>> On Tue, 5 Feb 2019, Richard Patterson wrote:
>> [...]
>>> If the device has permanent storage then it could ask for the previous
>>> PD it had,
>>
>> Yes, it could be done with a DHCPv6 message called 'CONFIRM' - it's sent
>> by the Client to the Server, and it 'confirms' the intention to use an
>> address that was allocated earlier.  It was specified for when like
>> waking up from standby mode.  It's specified for allocating Addresses,
>> so it would work the same when specified for prefixes.
>>
>> Alex
> 
> It could also just include it as a hint in the first SOLICIT.
> 
> Storing the old lease to persistent storage also gives the CPE
> visibility if the prefix does change.  If it does, it can then send
> RAs with lifetime 0 to deprecate the old prefix on the LAN side.

Which lifetime?  Prefix Information Option's (PIO's) Valid Lifetime or 
Preferred Lifetime?

I doubt this deprecation work straightforwardly on stacks of user's 
laptops (Windows) on LAN side, unless one tries, of course.

I am saying it because recently confronted with linux absence of visible 
lifetime fields in Neighbor Cache, otherwise urgently needed for other 
contexts.

Alex